THE PROBLEM OF DEFINING AND RECOGNIZING 

 SUBPOPULATIONS OF FISHES 



By 



John C . Marr— 



INTRODUCTION 



In trying to identify and measure the 

 causes of fluctuations in the abundance and dis- 

 tribution of a species of fish, it is essential 

 that the number and identity of subpopulations, 

 if any, within the species be established, since 

 each subpopulation may have its own character- 

 istic distribution, fecundity, natural mortality 

 rate, growth rate. etc. This statement (or 

 others li.<e it) is practically axiomatic in the 

 field of fishery biology, and yet, there is some 

 misunderstanding arising in part from semantic 

 difficulties and in part from the lacK of agreed 

 definitions of problems. 



The semantic difficulties are those of the 

 same term being used with different meanings 

 by different workers, different terms being used 

 with the same meaning, etc., often with no defin- 

 itions given. The terms which I will use may be 

 defined as follows: 



Population : A population of fish includes 

 all individuals of a given species when there 

 are no subspecies or, if there are subspecies, 

 when their distributions are not discrete. It in- 

 cludes only all individuals of a subspecies when 

 the distributions of the subspecies are discrete. 

 Obviously, there is gene flow, or opportunity 

 for such, throughout a population. 



Subpopulation : A subpopulation is a frac- 

 tion of a population that is itself genetically 

 self-sustaining. It is the smallest natural self- 

 perpetuating unit and is synonymous with the 

 term "deme" as used by systematists (Huxley, 

 Evolution the modern synthesis, New York, 1942: 

 203). Although differences between subpopula- 

 tions may be small, they are heritable . 



1/ Chief, South Pacific Fishery Investigations, 

 U.S. Fish and Wildlife Service, P.O. Box 271, 

 La Jolla, California . 



Stock: A stock is a population or a por- 

 tion of a population, all members of which are 

 characterized by similarities which are not 

 heritable, but are induced by the environment. 

 A stock may or may not include members of 

 several different subpopulations . 



Group: A group is a fraction of a popula- 

 tion with distinctive characteristics, the nature 

 of which (phenotypic or genotypic) has not yet 

 been determined. 



Race: I prefer not to use this term, 

 since in other fields (ornithology, for example) 

 it is a synonym of subspecies. The categories 

 involved in fishery problems are generally of 

 lesser rank than a subspecies; if not, they should 

 be recognized as such . 



Although it might be desirable to stand- 

 ardize on this or some other set of terms, this 

 is not essential so long as everyone defines 

 whatever terms he does use. 



The important distinction to keep in mind 

 between subpopulation and stock is that members 

 of a subpopulation segregate at spawning time, 

 whereas members of a stock need not. It is also 

 important to remember that members of a group 

 actually are members of either a subpopulation 

 or a stock, but cannot be assigned to one or the 

 other category because of lack of information. 

 The partial barriers to gene flow between sub- 

 populations are of the same nature as those be- 

 tween subspecies and include isolation in space, 

 isolation in time and ecological isolation. 



The "subpopulation problem, " then, in- 

 volves first, the recognition and definition of 

 each subpopulation and second, the planning of 

 biological observations such that the data accrue 

 discretely for each subpopulation rather than by 

 groups of subpopulations. If observations are 

 planned in such a manner that data accrue 



